WORKING PAPER 169/2018

Zareena Begum Irfan

Divya Jain

Satarupa Rakshit

Ashwin Ram

Modelling the Characteristics of Residential Energy Consumption: Empirical Evidence of

Indian Scenario

MADRAS SCHOOL OF ECONOMICS Gandhi Mandapam Road

Chennai 600 025

India

April 2018

MSE Working Papers

Recent Issues

* Working Paper 160/2017 An Alternative Argument of Green Solow Model in Developing Economy Context Santosh K. Sahu and Arjun Shatrunjay

* Working Paper 161/2017 Technical Efficiency of Agricultural Production in India: Evidence from REDS Survey Santosh K. Sahu and Arjun Shatrunjay

* Working Paper 162/2017 Does Weather Sensitivity of Rice Yield Vary Across Regions? Evidence from Eastern and Southern India Anubhab Pattanayak and K. S. Kavi Kumar

* Working Paper 163/2017 Cost of Land Degradation in India P. Dayakar

* Working Paper 164/2017 Microfinance and Women Empowerment- Empirical Evidence from the Indian States Saravanan and Devi Prasad DASH

* Working Paper 165/2017 Financial Inclusion, Information and Communication Technology Diffusion and Economic Growth: A Panel Data Analysis Amrita Chatterjee and Nitigya Anand

* Working Paper 166/2017 Task Force on Improving Employment Data - A Critique T.N. Srinivasan

* Working Paper 167/2017 Predictors of Age-Specific Childhood Mortality in India G. Naline, Brinda Viswanathan

* Working Paper 168/2017 Calendar Anomaly and the Degree of Market Inefficiency of Bitcoin S. Raja Sethu Durai, Sunil Paul

* Working papers are downloadable from MSE website http://www.mse.ac.in $ Restricted circulation

i

Modelling the Characteristics of Residential Energy Consumption:

Empirical Evidence of Indian Scenario

Zareena Begum Irfan Associate Professor, Madras School of Economics

(Corresponding Author) zareena@mse.ac.in

Divya Jain, Satarupa Rakshit and Ashwin Ram Madras School of Economics

ii

WORKING PAPER 169/2018

April 2018

Price : Rs. 35

MADRAS SCHOOL OF ECONOMICS

Gandhi Mandapam Road Chennai 600 025

India

Phone: 2230 0304/2230 0307/2235 2157

Fax : 2235 4847/2235 2155

Email : info@mse.ac.in

Website: www.mse.ac.in

iii

Modelling the Characteristics of Residential Energy Consumption: Empirical Evidence of Indian Scenario

Zareena Begum Irfan, Divya Jain, Satarupa Rakshit and Ashwin Ram

Abstract

Due to rapid economic expansion, India has one of the world's fastest growing energy markets and is expected to be the second-largest contributor to the increase in global energy demand by 2035, accounting for 18% of the rise in global energy consumption. Household sector is one of the largest users of energy in India, counting for about 30 per cent of final energy consumption (excluding energy used for transport) reflecting the importance of that sector in total national energy scenario. The pattern of household energy consumption represents the status of welfare as well as the stage of economic development. Household energy consumption is expected to increase in future along with growth in economy and rise in per capita incomes. This paper analysis the manifold aspects of anthropogenic activities with focus on household characteristics influencing the residential carbon dioxide emissions. Data on expenditure incurred for purchase used for both indoor and outdoor use as well as the socio-economic indicators have been taken into account to estimate their contribution to the level of emissions. The various indicators of household expenditure have been categorized separately under technological innovations, affluence, household demographics, biophysical characteristics and control variables. The theories of Ecological Modernization, Political Economy and Human Ecology have been highlighted to discuss the significance of household energy consumption pattern. Household income in India has increased considerably in line with economic growth over the last decades. In this study, household data has been extracted from India Human Development Survey- II (IHDS – II), 2011- 12 which is a cross sectional survey conducted by ICPSR 36151. The expenditures on energy consumption and emission factors are to a great extent influenced by the factors that are beyond the control of humans such as development trends regarding urban forms and infrastructure. In this study we have used household data which makes it an expanding literature to anthropogenic environmental degradation. Key words: urbanization, carbon emission, ecological modernization, energy consumption, economic expansion, human ecology, political economy

JEL Codes: C01, O31, 032, Q4, Q55, Q18, R21

iv

Acknowledgement The authors are grateful to their parent institute which provided them the infrastructural benefit of conducting the research work. The authors are also thankful to the audience and eminent researchers of Nanyang Technological University, Singapore and other research forums for providing their valuable suggestions during the East Asian Association of Environmental and Resource Economics, Annual Congress, 2017.

Zareena Begum Irfan Divya Jain

Satarupa Rakshit Ashwin Ram

1

INTRODUCTION

Earth’s global surface temperature is warming at a steady rate. It has

been rising for 11 years now between 1995 and 2006 and continues to

rise. Global warming is caused largely by the rising concentration of

GHGs (principally CO2, CH4 and N2O) in the earth's atmosphere, which

helps trap excess heat. The main reason behind this are the human

driven increase in the greenhouse gas emissions particularly carbon

dioxide which has posed a greatest challenge to climate change

problem. In its Fourth Assessment Report, the IPCC characterized

climate change as “any change in climate over time, whether due to

natural variability or as a result of human activity” (IPCC, 2007, 2013,

2014). Households remain one of the most significant contributors to

the climate change issue with some of the member states having thrice

or more emission footprint than the other states. Goods and services

consumed by humans induce pressures on the food and resources

needed to support human life. Moreover, building cities and constructing

roads remove a similar amount of bio productive land area. Carbon

footprint also explains the environmental impacts of human activities.

The definition of carbon footprint is the total CO2 emitted directly or

indirectly by a certain activity or the accumulation of CO2 during a

product life cycle.

There is a big difference in world’s per capita energy

consumption. The major difference is observed in the patterns of energy

consumption between developed and developing countries. The

developing nations like India and China are among the poor countries

seeking to develop its resources. Rapid economic expansion has lead

India has one of the world's fastest growing energy markets and is

expected to be the second-largest contributor to the increase in global

energy demand by 2035, accounting for 18% of the rise in global energy

consumption. Household sector is one of the largest users of energy in

India, accounting for about 30 per cent of final energy consumption

2

(excluding energy used for transport) reflecting the importance of that

sector in total national energy scenario (Reddy, 2003). The pattern of

household energy consumption represents the status of welfare as well

as the stage of economic development.

As the different economies develop, they consume higher

magnitude of energy and there is a rise in cleaner options too.

Household energy consumption is expected to increase in future along

with growth in economy and rise in per capita incomes. The projected

increases in household energy consumption are expected to result from

changes in lifestyles.

Household income in India has increased considerably in line

with economic growth over the last decades. The ministry of statistics

and programme implementation (MOSPI) reports that urban wages have

been rising by 17.38 % between 2000 and 2005.In line with wages also

household expenditure has been rising especially in the urban areas

were richer households are located. We expect a large share of

households to pass the critical income level of 2 dollars per day and we

expect that carbon emissions from Indian households will account for a

significant share of global greenhouse gas emissions (GHG) in the

future. The rise in carbon emissions is correlated with increasing direct

and indirect energy requirements of households. Energy consumption

and carbon footprints vary with what and how households consume. We

can account for the differences also on the basis of rural and urban

areas. Rural households spend their income on more energy intensive

commodities than a person from a metropolitan area on average. The

motivation behind this was to characterize household consumption

patterns with respect to their environmental implications and hereby

search for evidence on the Environmental Kuznets Curve (EKC). The

findings support previous research in the EKC energy literature, as

energy requirements increase monotonically with household expenditure

with no turning point observed (Stern, 2004).

3

There is still lack of knowledge on how energy consumption

relates to the demographic and economic characteristics of the

household (Brounen et. al., 2012), on the relative importance of

these characteristics, and whether changes in household's socio-

economic circumstances translate in changes in energy consumption.

The literature has analyzed how various individual, household and

housing characteristics correlate with environmental impacts, however

measured (Buchs and Schnepf, 2013b; Buchs et. al., 2011;

Tukker et. al., 2010). Many studies have used input output analysis

along with household consumption expenditure data in India.

Only few studies use multivariate regressions to analyze the

relative importance of the various socio-economic factors, and most of

them use cross-section data. This means that they cannot give us any

indication of what changes in household circumstances may have an

impact on energy consumption.

The objective of this paper was to estimate the impact of

anthropogenic activities particularly the household consumption

expenditure undertaken for both indoor and outdoor use on their

contribution to the overall emissions. The aim was to identify which

household characteristics and which changes in these characteristics

have the largest impact on energy expenditures. Such knowledge is

necessary to be able to design more effective policies to reduce the

carbon footprint of a country.

METHODOLOGY

Human dimensions of climate change phenomenon have become the

widely researched subject. The main domains which are included here

are the anthropogenic drivers of climate change, the effects of climate

change on ecosystems, mitigation or efforts to curb the emissions of

greenhouse gases and how do we adapt to the impacts of climate

change.

4

Three theoretical perspectives have been considered here include

ecological modernization (EM), political economy (PE) and Human

Ecology (HE). Ecological perspective is seen as the ecological

restructuring of processes of production and consumption in modern

societies due to environmental considerations. The reason is

modernization will lead to greater concern for environmental problems

among experts and this will spur the institutions to take steps to curb

environmental problems. The aspect of technological innovation is

considered to be an important factor underlying EM perspective

(Spaargaren and Mol, 2009; Foster et. al., 2010). It focuses on the

EKC phenomenon as discussed above also that first environmental

impacts increase in the early stages of development and falls once the

nation reaches the aggregate level of affluence.

This perspective stands in contrast with the political economy

perspective which says that rising level of affluence will lead to increasing

consumption and thus rising environmental impacts. Thus, the rising

portion of EKC is in line with the PE and the inverted U shaped curve

supports EM.

Human ecology researchers consider humans as part of the

ecosystem. Thus, despite their ingenuity and culture, humans are not

exempt from ecological constraints (Catton and Dunlap, 1980;

Dunlap, 1980, 1982). The HE perspective recognizes that population

characteristics (size, rate of growth, density, age structure) and affluence

are important drivers of anthropogenic environmental impacts. The

question addressed here is whether the biophysical characteristics play a

role in influencing anthropogenic environmental impacts. We apply the

HE perspective to examine the impacts of household-level demographic

characteristics and affluence on residential CO2 emissions.

In this study, household data has been extracted from India

Human Development Survey- II (IHDS – II), 2011- 12 which is a cross

5

sectional survey conducted by ICPSR 36151. Only IHDS survey variables

have been selected here which show the household level consumption

patterns depicting emission factors. Any other source which includes

household characteristics at state level has not been taken for the

present study. In order to see the impact of household characteristics on

indoor and outdoor household expenditure, we run the following two

multiple regressions-

Regression 1-

Here the dependent variable includes the expenditure on

consumption of LPG, kerosene, coal, household fuel and household

electricity.

= + + + + + +

+ + + + +

+

= Indoor Expenditure done by the household

= Total consumption expenditure

= Income

= Education of head of household

= Female education

= Religion for i = 1,2,3,4, 5

= Hours of access to electricity

For i = 0 if = 0

= 1 If 0 < < 6

= 2 if 5 < < 11

= 3 if 10 < < 16

= 4 if 15 < < 20

= 5 if 20 < < 25

= Hours of burning stove for i =0, 1, 2 …. 8…16

= Years in place for i = 1 …. 49

6

= 1 for household owns a refrigerator

0 if otherwise

= 1 if household owns a electric fan

0, otherwise

= 1 if household owns a generator set

0, otherwise

= 1 if household owns a air conditioner

0, otherwise

= 1 if household owns microwave oven

0, otherwise

= 1 if the household has window in the cooking area

0 if otherwise

= cooking place in the household if i = 1, 2, 3

Regression 2 –

Dependent Variable- Household Expenditure (Indoor) It has been

calculated by adding together the expenditure on the purchase of

personal transport equipment, spending on public transportation like

bus, auto, plane and Diesel, petrol and CNG maintenance.

= + + + + + + +

Here all the variables remain the same except for

= Household outdoor expenditure

= 1 if household owns a car

0 if otherwise

= 1 if household owns a motorcycle

0 if otherwise

7

Independent Variables:

Affluence-

In our model, we have considered household income and household

consumption expenditure under affluence. Affluence indicates the

household’s ability to spend which is an important component of its

income. The first indicator of affluence is household income.

Household income is analogous to GDP per capita, often used as a

measure of affluence in cross-national analysis of anthropogenic

environmental impacts. The EM perspective suggests income may have

an inverted U-shaped relationship with CO2 emission at the national

level, and suggests we should expect this type of relationship at the

individual or household level as well: Expenditure incurred both indoor

and outdoor should rise giving rise to CO2 emissions as income goes up

until a turning point is reached, after which they should decline with

further increases in income.

The second indicator which is household consumption

expenditure which shows the total value of goods and services

consumed. This should have a positive impact on the indoor and outdoor

household expenditure adding to the higher anthropogenic impacts of

carbon dioxide emissions.

Biophysical Characteristics-

Here the indicators used for biophysical characteristics are whether the

household has window in the cooking area and where the cooking is

done in the household. Dummies are assigned for both the indicators.

D1 = 1 for household has window in the cooking area

= 0, otherwise

D2 = 1 if cooking is done in the kitchen

D3 = 2 if cooking is done in the living area

8

Household Demographics-

The human ecology perspective of climate change regards the population

and its characteristics as important drivers of anthropogenic impacts.

Under household demographics, we consider firstly years in place taken

to be continuous variable. This is expected to have a positive impact on

the household expenditure resulting in a rise in the amount of carbon

dioxide emissions.

The second indicator pertains to the hours of access to

electricity. This will certainly increase the household expenditure due to

more energy consumption resulting in residential carbon dioxide

emissions.

The number of hours used by the household for burning stove

(cooking, heating water, and making tea) is expected to be positively

related to indoor carbon dioxide emissions. With the increase in the

number of hours of burning stove leads to increase in the emissions.

Whether the household owns a car or motorcycle as a part of its

personal transport equipment adds to the outdoor emissions due to the

anthropogenic activity. Therefore, as taken in every case dummy is

assigned to the two variables.

Control Variables-

We control for the effects of the variables which are not expected to

directly influence the dependent variable but have an impact on it in

some way or the other. Education of head of the household and the

female education impact the household emissions both indoor and

outdoor. More educated the head of the family or the more literate the

female in the household is, it is expected that they will take the

necessary measures that will decrease the emissions in terms of using

energy efficient technology.

9

One more variable considered here is religion. The relevance of

this variable to affect the household indoor and outdoor expenditure and

thus emissions is highly questionable. It is expected that people

belonging to different religions behave differently when undertaking their

expenditures which is not certainly true.

Technological Innovations-

Electric fan, refrigerator, microwave, air conditioner, generator set are

purchased by households or not. Dummies are assigned to the five

different categories. They form an important part of ecological

perspective which says that innovations will decrease the amount of

emissions with the effect being negative. Thus, the owning capacity tells

us about their effect on the carbon emissions.

Descriptive statistics for all the variables is analyzed in Table 1.

Table 1: Descriptive Statistics of Dependent and Independent Variables

MEAN

STANDARD DEVIATION

RANGE

MIN MAX

Dependent Variable 1.Household Expenditure(INDOOR)

1434.343

2180.567

90

89558

2.HOUSEHOLD Expenditure(OUTDOOR)

2274.377 16616.32 0 1202400

Technological

Innovations:

Whether household owns-

1..Electric Fan

2. Fridge / Refrigerator

3. Generator set

4. Microwave Oven

5. Air conditioner

.7543237

.2750947

.0181235

.0167818

.0206521

.4304927

.4465674

.1333997

.1284545

.1422185

0

0

0

0

0

1

1

1

1

1

10

MEAN

STANDARD DEVIATION

RANGE

MIN MAX

Affluence: 1.Household Consumption

Expenditure 2.Household Income

117484.6

126611.2

120048.9

216516.4

180

-1037040

4080760

1.14e+07

Household

Demographics: 1.Years in place (1-46)

2.Owning of

vehicles(Yes=1) a. Car

b. Motorcycle 3.hours of access to

electricity

77.29898

.0480045

.2864326

.6403651

25.77968

.2137783

.4520995

.4798992

0

0 0

0

90

1 1

1

Biophysical Characteristics: 1) Having a window or vent in cooking area?

2)Cooking place in the house a. Separate Kitchen

b. Living area

.7518739

.584798 .2365618

.4319321

.4927628 4249761

0

0 0

1

1 1

Control Variables: 1. Education of head of

household 2. Female education

a.1

b.3 c.4

3. Religion a.2

b.3

c.4

.221024

.5389963

.0731365

.0196707

.1208891

.0232129

.0236067

.4149422

.4984831

.2603636

.1388676

.3260024

.1505809

.1518223

0

0

0 0

0

0

0

1

1

1 1

1

1

1

Data Interpretation and Discussion

The graphical analysis has been shown below by calculating the average

for different states as well for India at aggregate level. States such as

Mizoram, Daman and Dui, Dadar and Nagar Haveli, Sikkim, Nagaland,

11

Manipur, Nagaland, Tripura, Meghalaya, Pondicherry and Manipur have

been excluded from the study due to some missing observations for

various variables in order to avoid ambiguous results.

12

The above three Figures 1-3 can be interpreted as follows-

New Delhi spends the highest on household electricity compared

to other states. Jammu and Kashmir, Chandigarh, New Delhi and Punjab

spend more than the Indian average on electricity. If we compare this

with number of hours household has access to electricity we can observe

that in spite of incurring significant expenditure on electricity, Jammu and

Kashmir has less number of electricity access hours. The main reason for

this could be climatic conditions prevailing there which in turn affect their

hours available for electricity use.

The same happens when the expenditures on the cooking and

household appliances is taken into account. Average household

expenditure on electricity for Jammu and Kashmir, New Delhi and

Chandigarh is positively related to their expenses on cooking and

household appliances which includes electric fan, AC, washing machine,

refrigerator etc.

13

TOTAL EXPENDITURE ON DIESEL, PETROL AND DIESEL

MAINTAINENCE DURING PAST ONE MONTH

Figure 4 and 5 show that almost all the states spend more than

the Indian average on the purchase of personal transport equipment with

Himachal Pradesh and Punjab spending the highest adding the most.

Chandigarh, Himachal Pradesh and Punjab spend less on Public transport

modes compared to private modes adding to more energy expenditures

and thus more emissions. In contrast to the above case, New Delhi

spends more on public modes of transport versus private mode. This

depicts higher expenditure on public transportation for India as a whole.

Thus, less are the outdoor emissions due to household activities.

14

In Figure 6, we can see that analogous to their expenditure

on purchase of personal transport equipment the same states that is

Chandigarh, Punjab and Jammu and Kashmir spend the highest on

diesel and petrol maintenance.

Figure 7 does not explicitly show difference in the expenditures done by

households in different regions in India to make possible comparisons.

15

Apart from a few states, almost all the states spend more than Indian

average on the purchase of different types of household fuel.

With most of the state spending more than the Indian average on the

purchase of LPG leads to its higher expenditure and thus, more

household indoor emissions (Figure 8).

16

Expenditure done on kerosene is highest by New Delhi as per the

Figure 9. New Delhi, Punjab and Chandigarh spend more than the

Indian average which shows higher Indian average expenditure on

kerosene adding to the indoor emissions.

17

Coal is either collected from the nearby village or some other

place or it is purchased. Jammu and Kashmir, Jharkhand and Karnataka

incur high amount of expenditures on the purchase of coal (Figure 10).

We will look at the next four figures to see the time taken by the

households in different States to collect fuel.

From the four figures (Figure 11-14), we can see that

Chhattisgarh spends highest time in collecting fuel. This applies to both

boys and girls under 15 years of age and adult women and adult men

more than 15 years of age. The reason behind this could be that

Chhattisgarh being a backward region tops the list of states in terms of

its poverty rate. So, people over there spend more amount of time in

collecting fuel rather than purchasing it. The economic condition of the

households may not be financially sound for them to purchase fuel.

18

19

RESULTS AND DISCUSSION

Regression results from the two regression equations stated previously

are discussed here. Similar to the graphical interpretation, here also the

various states have not been taken into consideration.

From the first regression, we obtain that the R square value

shows that 14.22 % of the variation in the indoor expenditure is

explained by the explanatory variables such as income, owning capacity

of household, hours of access to electricity and hours of burning stove or

chulha. The R square value obtained above is pretty good for a cross

sectional surveys like this.

Education of the head of the household given by headedu has

insignificant p value for the both the dummies generated for the three

categories. Cototal ie total consumption expenditure for all the purposes

has p value as 0.000 which is highly significant and seems to be

20

positively related to the indoor expenditure. When the total household

consumption expenditure increases by 1 unit, the increase in the

household indoor expenditure is by 0.0047 units.

The variables such as income and years in place have p values

greater than 0.1 which depicts their insignificant effects on the indoor

expenditure. Biophysical characteristics explained by the cooking place in

the house and whether the household has a window or vent in the

cooking area have insignificant p values.

Hours of burning stove or chulha has a positive effect on the

indoor expenditure contributing to higher indoor emissions. This is

supported in the above results as their p values turns out to be

significant. With the increase in the number of burning hours of stove

from 1-8 hours by 1 unit, the indoor emissions rise due to the rise in the

indoor expenditure from 167.07 to 673.27.

Female education and religion do not have a significant effect on

the household indoor expenditure as the null hypothesis is accepted at

10% level of significance due to their p values greater than 0.1.

Number of hours the household has access to electricity has p

value 0.14 which is slightly greater than 0.1. Hence we can say that the

null hypothesis for such a variable can be rejected to show its significant

effect on the indoor expenditure and thus emissions. The increase in the

number of hours the electricity access by 1 unit leads to increase in the

indoor expenditure by 89.92 units. This has a overall effect on rising

household indoor emissions.

Owning capacity of households in India for indoor use is shown

by whether the household owns a generator set (cg6), refrigerator

(cg18), electric fan (cg13), microwave (cg27), air conditioner (cg22). For

all the variables, p value is significant except for cg6. cg13 which shows

whether the household owns a electric fan or not. The average

21

expenditure when the household has a electric fan is greater by 244.79

units as compared to the base category. cg18 which holds for refrigerator

shows that the average expenditure for the household on refrigerator is

higher by 324.544 units when compared with the benchmark category.

Similar interpretation holds for cg27 as the average expenditure on

microwave oven is more by 478.98 units. The average expenditure

incurred by the household when it owns a refrigerator is greater by

891.38 units as compared to when it does not.

On examining the p values shows us the significance for each of the

variables. At 10 % level of significance, the p value ie the probability of

committing type I error shows that if p > 0.1, then we accept the null

hypothesis whereas if p < 0.1 we reject null hypothesis. We see that

almost all the variables are significant over here as their p values are

turning out to be less than 0.1.

R square for this regression comes out to be 11% which means

that 11% of the variation in the expenditure is explained by the all the

independent variable including income, years in place, whether the

household owns motor cycle or car.

Years in place represented by id15 are highly significant as

represented by its p value and has a positive relationship with outdoor

expenditure. If number of years that a household has stayed in a place

increases by 1 unit, their expenditure increases by 14.73 units.

P value of Total consumption expenditure (cototal) is given by

0.00 is significant at 10 % level of significance. It is as expected to be

positively related with the household expenditure. If the consumption

expenditure increases by 1 unit, then the outdoor expenditure increases

by 0.04 units. As the income of the household increases, the expenditure

decreases by 0.011 units. The p value for the income variable also turns

out to be significant.

22

Education of head of family given by headedu is being divided

into three categories. For three categories we generate two dummies for

1 and 2. 3 is taken as the benchmark category. Higher the education of

head of the family, the less will be the household outdoor expenditure for

the activities contributing to the emissions. P value for the variable turns

out to be insignificant.

This is taken as a part of the control variable which may or may

not have an effect on the expenditure undertaken.

Cg8 and cg21 explain whether the household owns motorcycle or

car for which we get significant p values as 0.00. The average

expenditure when the household owns the motorcycle is more by

1666.554 as compared to when the household does not own a

motorcycle. Similarly, the average outdoor expenditure when the

household owns a car is more by 5885.127 as compared to the base

category. For the Female education also, the p value is rejected at 10 %

level of significance as it is less than 10% level of significance. More the

females are educated in the household, the less will be expenditure

incurred by them on the activities contributing to emissions. There

appears to be a negative relationship between female education and

expenditure incurred for outdoor purposes. If the female education

increases by 1 unit, expenditure falls by 527.944.

Religion explained by three dummies. The base category belongs

to the people who are Hindu. P value for the people who are Muslim is

significant and expenditure for them is lower by 663.04 as compared to

the outdoor expenditure by Hindus. P values for the other two dummies

are not significant as we accept the null hypothesis.

The characteristic such as whether the household has window in

the cooking area or where cooking is done in the household have their p

23

values as significant. The average expenditure done by the household for

outdoor purposes if it has window in the cooking area is less than the

base category by 592.73 units. The differences between the states with

almost all displaying significant p values at 10% level of significance. We

see the behavior of variables for India as a whole rather than for

different states separately.

Summary of the Findings:

We summarize our findings in the form of a table in relation to the

theories evaluated- EM, PE, HE.

Table 2 provides the information on the relationship between

each of the independent variables and dependent variables of household

expenditure. It also provides information on the theory (theories)

supported by clarifying remarks for each of them. When we analyse EM

perspective for Indian households we find that the aspect of

technological innovations is not in line with the theory as it has a positive

effect on expenditure and thus increasing emissions.

They support PE perspective due to the plus sign which prevails.

Household income has a negative effect on the outdoor expenditure and

thus follows the point stated by EM theorists regarding the EKC curve.

Household demographics play a key role in driving anthropogenic

environmental impacts and both biophysical characteristics combined do

not give a clear indication of their impact on the dependent variable.

Thus, HE perspective shows that household demographic characteristics

are important drivers to anthropogenic carbon emissions.

24

NOTE: EM (Ecological Modernization), HE (Human Ecology), PE (Political Economy) NS= Not significant, (-) Negative Significant Relationship, (+) Positive Significant Relationship, # = Theory or Theories supported by each statistically significant finding,

25

For dummy- coded variables with multiple categories (home insulation and type of glass in windows) we present a summary for the variable as a whole rather than for each dummy-coded category, where a significant effect in at least one category is noted.

EM does not necessarily predict a monotonically negative

relationship, but it suggests that the relationship should become negative

at higher levels of affluence.

CONCLUSIONS

The issue of climate change caused by the residential anthropogenic

drivers such as household demographics, biophysical characteristics,

affluence and technological innovations has been discussed in the present

paper. The diagrammatic representation for the expenditure patterns

reveals three main points from the survey for the year 2011-12.

- Households in Jammu and Kashmir spend the highest on electricity

and on the purchase of cooking and household appliances

contributing to the highest energy expenditures which poses higher

indoor emission burden on India.

- In comparing the expenditure on the private means of transportation

with public modes, we observe that New Delhi spends more on the

use of public modes and households in Chandigarh spend more on

private mode. Thus, New Delhi adds less and Chandigarh adds to

more outdoor emissions as a part of India’s GHG emissions

trajectory.

- Time taken by females and males in collection of fuel is highest by

Chhattisgarh due to the suitable reasons such as backwardness of

the region which stands in contrast with other places in India.

The regression results show somewhat the significance of the

various variables as well as their impact on the two dependent variables

from the two regressions by looking at their respective p values and the

26

relationship shown by the sign respectively. Main results come out to be

as follows-

- Technological innovations such as owning of electric fan, refrigerator,

microwave oven, air conditioner which have been considered here

have significant positive effects on indoor expenditure except for

generator set which has insignificant effect. This stands in line with

the PE theory but contradicts EM perspective which regards

technological innovation as a holy grail of human efforts to mitigate

the adverse environmental consequences of modern economic

growth.

- Affluent indicators such as household consumption expenditure

shows positive effect on both the expenditures which confirms to the

PE theorists who say that green technologies produce environmental

benefits but they are counteracted by affluence based increased

consumption. Household income has a negative effect on the

expenditure indicating that as India is on the path to development

the environmental degradation has started to decline with the rise in

income. Since, the analysis is done only for one period over here we

cannot make definite conclusions.

- Household demographics and biophysical characteristics also play

an important part in explaining issue of anthropogenic environmental

degradation. Most of the household demographic factors positively

affect the expenses in a positive way. On the other hand, biophysical

characteristics do not give a clear indication of their effects on the

carbon dioxide emissions .Hence; HE theory shows that affluence

and household demographics drive anthropogenic carbon dioxide

emissions.

It is crucial to recognize the expenditures on energy consumption and

emission factors are to a great extent influenced by the factors that are

beyond the control of humans such as development trends regarding

urban forms, housing density, type and infrastructure. The policies

which affect urbanization and urban development play important roles in

27

determining household energy use as well as some of political economic

factors. These issues which are extremely essential have not been taken

into account while carrying out the study. Moreover, we could not

account for the difference between the rural and urban areas within

India which could have made our analysis much simpler. This was not

possible due to less information available for all the states and different

sources did not permit us to combine them into one dataset as the

analysis was done for 2011-2012. It can be seen that many studies

which have been conducted earlier have considered set of predictors

which was not enough. But in our study we have used household data

this makes it an expanding literature to anthropogenic environmental

degradation.

REFERENCES

Catton, Jr. W and Dunlap, R.E. (1980), “A New Ecological Paradigm for Post-Exuberant Sociology”, American Behavioral Scientist, 24,15-

47.

Dunlap, R.E. (1980), “Paradigmatic Change in Social Science: from

Human Exemptionalism to an Ecological Paradigm”, American Behavioral Scientist, 24, 5-14.

Foster, J.B., B. Clark and R. York (2010), “The Ecological Rift:

Capitalism's War on the Earth”, Monthly Review Press, New York.

IPCC (Intergovernmental Panel on Climate Change) (2007), “An

Assessment of the Intergovernmental Panel on Climate Change”, Climate Change 2007: Synthesis Report

IPCC (2013), “Climate Change 2013: the Physical Science Basis”

Summary for Policymakers.

IPCC (2014), “Climate Change Synthesis Report”, IPCC Fifth Assessment Synthesis Report.

Reddy, S.B. (2003), ”Overcoming the Energy Efficiency Gap in India's

Residential Sector”, Energy Policy, 31(11).

28

Spaargaren,G and Mol,A.P.J. (2009), “Sociology, Environment, and

Modernity: Ecological Modernization as a Theory of Social Change”, A.P.J. Mol, D.A. Sonnenfeld,G. Spaargaren(Eds.), The Ecological Modernization Reader: Environmental Reform in Theory and Practice, Routledge, London and New York (2009), 56-79.

Stern, I.D. (2004), “The Rise and Fall of the Environmental Kuznets Curve”, World Development, 32, 1419-1439.

MSE Monographs

* Monograph 30/2014

Counting The Poor: Measurement And Other Issues

C. Rangarajan and S. Mahendra Dev

* Monograph 31/2015

Technology and Economy for National Development: Technology Leads to Nonlinear

Growth

Dr. A. P. J. Abdul Kalam, Former President of India

* Monograph 32/2015

India and the International Financial System

Raghuram Rajan

* Monograph 33/2015

Fourteenth Finance Commission: Continuity, Change and Way Forward

Y.V. Reddy

* Monograph 34/2015

Farm Production Diversity, Household Dietary Diversity and Women’s BMI: A Study of

Rural Indian Farm Households

Brinda Viswanathan

* Monograph 35/2016

Valuation of Coastal and Marine Ecosystem Services in India: Macro Assessment

K. S. Kavi Kumar, Lavanya Ravikanth Anneboina, Ramachandra Bhatta, P. Naren,

Megha Nath, Abhijit Sharan, Pranab Mukhopadhyay, Santadas Ghosh,

Vanessa da Costa and Sulochana Pednekar

* Monograph 36/2017

Underlying Drivers of India’s Potential Growth

C.Rangarajan and D.K. Srivastava

* Monograph 37/2018

India: The Need for Good Macro Policies (4th Dr. Raja J. Chelliah Memorial Lecture)

Ashok K. Lahiri

* Monograph 38/2018

Finances of Tamil Nadu Government

K R Shanmugam

* Monograph 39/2018

Growth Dynamics of Tamil Nadu Economy

K R Shanmugam

WORKING PAPER 169/2018

Zareena Begum Irfan

Divya Jain

Satarupa Rakshit

Ashwin Ram

Modelling the Characteristics of Residential Energy Consumption: Empirical Evidence of

Indian Scenario

MADRAS SCHOOL OF ECONOMICS Gandhi Mandapam Road

Chennai 600 025

India

April 2018

MSE Working Papers

Recent Issues

* Working Paper 160/2017 An Alternative Argument of Green Solow Model in Developing Economy Context Santosh K. Sahu and Arjun Shatrunjay

* Working Paper 161/2017 Technical Efficiency of Agricultural Production in India: Evidence from REDS Survey Santosh K. Sahu and Arjun Shatrunjay

* Working Paper 162/2017 Does Weather Sensitivity of Rice Yield Vary Across Regions? Evidence from Eastern and Southern India Anubhab Pattanayak and K. S. Kavi Kumar

* Working Paper 163/2017 Cost of Land Degradation in India P. Dayakar

* Working Paper 164/2017 Microfinance and Women Empowerment- Empirical Evidence from the Indian States Saravanan and Devi Prasad DASH

* Working Paper 165/2017 Financial Inclusion, Information and Communication Technology Diffusion and Economic Growth: A Panel Data Analysis Amrita Chatterjee and Nitigya Anand

* Working Paper 166/2017 Task Force on Improving Employment Data - A Critique T.N. Srinivasan

* Working Paper 167/2017 Predictors of Age-Specific Childhood Mortality in India G. Naline, Brinda Viswanathan

* Working Paper 168/2017 Calendar Anomaly and the Degree of Market Inefficiency of Bitcoin S. Raja Sethu Durai, Sunil Paul

* Working papers are downloadable from MSE website http://www.mse.ac.in $ Restricted circulation